基于推荐算法的高血压患者偏好方法研究  

Research on Preference Method of Hypertensive Patients Based on Recommendation Algorithm

作　　者：赵贝宁 熊巨洋[1] 赵仁伟 金唐 ZHAO Beining;XIONG Juyang;ZHAO Renwei;JIN Tang(School of Medicine and Health Management,Huazhong University of Science and Technology,Wuhan 430030;Central School of Guoyuan Township in Mianchi County,Mianchi 472400)

机构地区：[1]华中科技大学医药卫生管理学院,武汉430030 [2]河南省渑池县果园乡中心学校,渑池472400

出　　处：《计算机与数字工程》2022年第5期978-982,共5页Computer & Digital Engineering

基　　金：国家自然科学基金面上项目“基于离散选择实验的我国慢性病患者中医药服务利用效用及供给策略研究”(编号:71673095)资助。

摘　　要：随着人们生活水平的提高,高血压患者呈现增多的态势,有效找到高压患者的健康管理偏好是帮助高血压患者快速改善身体状况的重要渠道。而协同过滤技术是当前推荐算法中效果最好的方法之一,但传统的协同过滤技术的具有实时性较差,扩展性较弱等缺点。针对这种情况,提出了基于协同过滤推荐算法的高血压患者偏好方法。首先,将健康管理方式项目聚类,并对项目进行评分预测,建立“高血压患者-健康管理”项目评分矩阵;其次,将健康管理项目聚类结果与高血压患者行为特征相结合,并对患者聚类;最后,利用近邻相似性计算实现对不同高血压患者健康管理偏好的推荐。通过实验分析结果可知,相对于传统的协同过滤推荐算法,该方法在高血压患者健康管理偏好推荐中具有准确性强、实时性强的特点。With the improvement of people's living standards,hypertensive patients show an increasing trend.Effectively finding the health management preferences of hypertensive patients is an important way to help hypertensive patients to quickly improve their physical conditions.The collaborative filtering technology is one of the best methods in the current recommended algorithm,but the traditional collaborative filtering technology has the disadvantages of poor real-time performance and weak scalability.In view of this situation,a preference method for hypertensive patients based on collaborative filtering recommendation algorithm is proposed.Firstly,health management items are clustered and the scoring of the items is predicted to establish a"hypertensive patient-health management item"scoring matrix.Secondly,the health management item clustering results are combined with the behavioral characteristics of hypertensive patients and the patients are clustered.Finally,nearest neighbor similarity calculation is used to implement recommendations for health management preferences of different hypertensive patients.According to the experimental analysis results,compared with the traditional collaborative filtering recommendation algorithm,this method has the characteristics of high accuracy and strong real-time in the recommendation of health management preference of hypertensive patients.

关 键 词：高血压患者 健康管理偏好 协同过滤技术 推荐算法 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]